KR101579530B1 - Vehicle, apparatus and method of controlling voice recognition of vehicle - Google Patents

Abstract

Disclosed are a vehicle, an apparatus and a method for controlling voice recognition of a vehicle. According to one aspect of the present invention, the objective is to register a new reservation word rapidly and simply with participation of a driver. According to another aspect of the present invention, another objective is to register a new reservation word rapidly and simply without participation of the driver. To this end, the method for controlling voice recognition of a vehicle according to the present invention includes the steps of: acquiring information on a new broadcasting channel for registration of the new broadcasting channel; acquiring a feature pattern of a voice signal generated by articulation of a name of the new broadcasting channel; and registering the feature pattern of the voice signal and the information of the new broadcasting channel for voice recognition control.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice recognition apparatus,

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle, and more particularly, to a vehicle based on voice recognition control.

Speech recognition is a series of processes in which phonemes, or linguistic information, are extracted from the acoustic information contained in the voice and are recognized and reacted by the machine. Speech conversation is perceived as the most natural and easy way to communicate among a large number of human and machine information exchanges, but in order to communicate with machine and voice, there is a restriction to convert the human voice into a code that can be processed by the machine . Speech recognition is the process of converting these codes.

In the automobile, a user interface based on voice recognition control is provided for the convenience of the user. In particular, audio-video / navigation (AVN) is also controlled based on voice recognition, so that the functions of AVN can be used without directly pressing a button or touching a touch screen.

In order to register a new reserved word in such a device based on speech recognition control, a new speech recognition database in which a new reserved word is registered needs to be updated to update the speech recognition control system. It is not desirable in terms of rapid customer service.

According to an aspect of the present invention, it is an object of the present invention to enable quick and easy registration of a new reserved word with participation of a driver.

According to another aspect of the present invention, there is another object to allow a new reserved word to be registered quickly and easily without the participation of a driver.

According to another aspect of the present invention, there is provided a method of controlling a voice recognition of an automobile, the method comprising: acquiring information of a new broadcast channel for registration of a new broadcast channel; Obtaining a feature pattern of a speech signal generated by utterance of a name of a new broadcast channel; And registering a new broadcast channel information and a feature pattern of the voice signal for voice recognition control.

Further, in the above-described voice recognition control method for a car, the step of acquiring a feature pattern of a voice signal may include: inducing a driver to utter a name of a new broadcast channel; And analyzing a voice signal generated by the driver's utterance to obtain a feature pattern.

Further, in the speech recognition control method of the above-described automobile, the speech inducing step includes the step of displaying a guidance message for inducing utterance of the name of the new broadcast channel of the driver.

Further, in the above-described voice recognition control method for a car, the presentation of the guidance message is to display the guidance text for inducing utterance of the name of the new broadcast channel through the display.

Further, in the above-described voice recognition control method for a car, the presentation of the guidance message is to output the announcement for inducing utterance of the name of the new broadcast channel through the speaker.

Further, in the above-described voice recognition control method for an automobile, the step of acquiring the characteristic pattern of the voice signal includes the steps of converting the character string of the name of the new broadcast channel into the voice signal using the character-voice conversion means; And analyzing the speech signal generated by the character-to-speech conversion to obtain a characteristic pattern.

Further, in the voice recognition control method of the automobile described above, the information of the new broadcast channel includes the name of the new broadcast channel and the service ID.

According to another aspect of the present invention, there is provided a vehicle including: a tuner unit for acquiring information on a new broadcast channel for registering a new broadcast channel; A speech recognition processor for obtaining a characteristic pattern of a speech signal from a speech signal generated by speech of a name of a new broadcast channel; And a controller for registering the information of the new broadcasting channel and the characteristic pattern of the voice signal for voice recognition control.

In the automobile described above, the speech recognition processing section induces the driver to ignite the name of the new broadcast channel, and analyzes the speech signal generated by the driver's speech to acquire the characteristic pattern.

Further, in the above-described automobile, the control unit controls the driver to display a guidance message for inducing utterance of the name of the new broadcast channel.

Further, in the automobile described above, the presentation of the guide message is to display the guide text for inducing utterance of the name of the new broadcast channel through the display.

Further, in the above-described automobile, the presentation of the guidance message is to output the announcement for inducing utterance of the name of the new broadcast channel through the speaker.

In the automobile described above, the speech recognition processor converts the character string of the name of the new broadcast channel into a speech signal by using the character-speech conversion means, analyzes the speech signal generated by the character-speech conversion, do.

In addition, in the automobile described above, the information of the new broadcasting channel includes the name of the new broadcasting channel and the service ID.

Another speech recognition control method of an automobile according to the present invention for acquiring a new broadcast channel includes: acquiring information of a new broadcast channel for registration of a new broadcast channel; Displaying a guide message to induce a name of a new broadcast channel to be ignited; Obtaining a feature pattern of a speech signal generated by utterance of a name of a new broadcast channel; And registering a new broadcast channel information and a feature pattern of the voice signal for voice recognition control.

Further, in another speech recognition control method of the automobile described above, the presentation of the guide message is to display the guide text for inducing utterance of the name of the new broadcast channel through the display.

Further, in another voice recognition control method of the automobile described above, the presentation of the guide message is to output a guide for inducing utterance of the name of the new broadcast channel through the speaker.

According to another aspect of the present invention, there is provided a vehicle including: a tuner unit for acquiring information on a new broadcast channel for registering a new broadcast channel; A speech recognition processor for obtaining a characteristic pattern of the speech signal from the speech signal; Controls the output unit to display a guidance message for inducing the name of the new broadcast channel to be displayed, controls the speech recognition processing unit to acquire the feature pattern of the speech signal generated by uttering the name of the new broadcast channel, And a control section for registering the characteristic pattern of the information and the voice signal in the voice recognition processing section.

Further, in another automobile described above, the presentation of the guide message is to display the guide text for inducing utterance of the name of the new broadcast channel through the display.

Further, in another automobile described above, the presentation of the guide message is to output the announcement for inducing utterance of the name of the new broadcast channel through the speaker.

Another speech recognition control method of an automobile according to the present invention for acquiring a new broadcast channel includes: acquiring information of a new broadcast channel for registration of a new broadcast channel; Converting a character string of a name of a new broadcast channel into a voice signal using a character-voice conversion means; Obtaining a characteristic pattern of a speech signal generated by the character-to-speech conversion; And registering a new broadcast channel information and a feature pattern of the voice signal for voice recognition control.

According to another aspect of the present invention, there is provided a vehicle including: a tuner unit for acquiring information on a new broadcast channel for registering a new broadcast channel; A speech recognition processor for converting a character string of the name of the new broadcast channel into a speech signal by using the character-to-speech conversion means and acquiring a characteristic pattern of the speech signal from the speech signal; And a control unit for registering the information of the new broadcasting channel and the characteristic pattern of the voice signal in the voice recognition processing unit.

According to an aspect of the present invention, it is possible to easily update the voice recognition control system by allowing a new reserved word to be quickly and easily registered with participation of a driver.

According to another aspect of the present invention, it is possible to quickly and easily register a new reserved word without participation of a driver, thereby facilitating update of the voice recognition control system while improving the convenience of the driver.

1 is a view showing the inside of a vehicle according to an embodiment of the present invention.
2 is a diagram showing a configuration of an AVN 100 for a car.
Fig. 3 shows a configuration of the speech recognition processing unit shown in Fig.
4 is a diagram illustrating a method of controlling speech recognition of a vehicle according to an embodiment of the present invention.
5 is a diagram illustrating a method of acquiring and registering information on a new broadcast channel for voice recognition control of an automobile according to an embodiment of the present invention.
6 is a diagram illustrating a first embodiment of a method of registering a new broadcast channel according to an embodiment of the present invention.
7 is a diagram illustrating a second embodiment of a method of registering a new broadcast channel according to an embodiment of the present invention.
8 is a diagram illustrating a method of selecting a broadcast channel through voice recognition after updating a new broadcast channel for voice recognition control of an automobile according to an embodiment of the present invention.

1 is a view showing the inside of a vehicle according to an embodiment of the present invention. 1, the AVN 100 and the steering wheel 102 are mounted on the front of the driver's seat. The AVN 100 is an integrated audio / video / navigation, and includes a display (see 214 in FIG. 2). The AVN 100 is based on speech recognition control. To this end, a voice recognition button 104 is mounted on the steering wheel 102. A microphone 106 is mounted on the upper portion of the driver's seat. A speaker 116 is mounted on the left door of the driver's seat and the right door of the accompanying magnet. The AVN 100 of the vehicle according to the embodiment of the present invention is based on voice recognition control and the voice recognition button 104, the microphone 106, the speaker 116 and the like are used for voice recognition control of the AVN 100 It can be used as an auxiliary tool.

2 is a diagram showing a configuration of an AVN 100 for a car. The AVN 100 shown in FIG. 2 is based on voice recognition control. As shown in Fig. 2, the AVN 100 is largely composed of elements for a voice recognition function, elements for general input functions, elements for broadcasting / communication functions, elements for navigation functions, Elements, and elements that can be used in common for a plurality of functions.

The configuration for the voice recognition function includes a voice recognition button 104 and a microphone 106, a voice recognition processing unit 208, and a command output interface 218. [ Elements for the broadcast / communication function include an antenna 252, a tuner unit 254, a broadcast signal processing unit 256, and a communication signal processing unit 258. The elements for the navigation function include a navigation database 262 and a navigation drive 264. [ The elements for the audio / video function include an audio / video input unit 272 and an audio / video playback unit 274. [ The configuration for a typical input function includes an input 272. Elements that can be commonly used for a plurality of functions include a memory 210 and a control unit 212, a display 214, and a speaker 116. Such functional division is not limited to those described above, and an element for one function may be used for another function.

The voice recognition button 104 allows the driver to use the audio function of the AVN 100 by executing a composite function such as a video function, a navigation function, and an information communication function. For this purpose, the voice recognition button 104 supports a one-key operation of a push-to-talk (PTT) scheme. The voice recognition button 104 may be installed on the steering wheel 102 so that the driver can operate the vehicle comfortably during operation. The steering wheel 102 is a steering device used to change the traveling direction of the automobile by moving the wheels of the automobile to the left and right. Since the driver grasps the steering wheel 102 all the time while the driver is driving, if the voice recognition button 104 is installed on the steering wheel 102, the driver can conveniently operate the voice recognition button 104 during operation. In addition to the steering wheel 102, if the driver is able to easily operate the voice recognition button 104 during operation, the voice recognition button 104 may be provided at any position of the vehicle.

The microphone 106 receives a voice signal generated by the driver while the voice recognition control function is being executed and converts the received voice signal into an electric signal. The microphone 106 may be a microphone provided for speech recognition control or a microphone for hands-free use of a car. Also, the microphone 106 may be a microphone of a mobile terminal carried by the driver. When a microphone of the mobile terminal is used, the mobile terminal and the AVN 100 must be connected to each other through Bluetooth or other short-distance communication.

The speech recognition processing unit 208 performs speech recognition on the electrical signal converted by the microphone 106 and extracts speech command information as a result of speech recognition. The voice command information extracted by the voice recognition processing unit 208 is transmitted to the control unit 212. [

The command output interface 218 is for transmitting a control command signal corresponding to voice command information extracted as a result of voice recognition from the control unit 212 to the control target apparatus.

The antenna 252 is a device for receiving a broadcast signal or for transmitting or receiving a communication signal, or for transmitting a radio wave to the air. The antenna 252 is communicably connected to the tuner unit 254. Therefore, the radio wave received by the antenna 252 is transmitted to the tuner unit 254. The antenna 252 may be composed of a plurality of types of antennas for a plurality of different types of broadcast / communication signals.

The tuner unit 254 receives the radio wave received by the antenna 252 and converts it into an intermediate frequency signal or the like. The tuner unit 254 converts the data signal to be transmitted into a form that can be propagated to the air, and sends the data signal to the air through the antenna 252. That is, the tuner unit 254 performs operations such as extracting only a signal of a specific band or combining a data signal with a carrier signal. The tuner unit 254 performs reception of broadcast signals and transmission and reception of communication signals. The broadcast signal may include a radio broadcast signal and a DMB (Digital Multimedia Broadcasting) broadcast signal. The communication signal may include a satellite communication signal with a Global Positioning System satellite (hereinafter, GPS satellite). The communication signal may also include a communication signal for telematics. Which signal is to be received and processed by the tuner unit 254 is determined by the control signal transmitted to the tuner unit 254 by the control unit 212. [ For example, when the control unit 212 generates a control signal to the tuner unit 254 so as to receive a radio broadcast signal of a specific channel, the tuner unit 254 outputs the radio signal of the corresponding channel to the tuner unit 254 in response to the control signal transmitted from the control unit 212. [ And receives broadcast signals. If the control unit 212 transmits a control signal and transmission data for transmitting the telematics signal to the tuner unit 254, the tuner unit 254 transmits the transmission data to the air in response to the control signal transmitted from the control unit 212 And transmits the converted signal through the antenna 252 to the air. Also, the tuner unit 254 acquires the information of the broadcast channel included in the broadcast signal. The broadcast signal input to the tuner unit 254 includes a name of a broadcast channel, a service ID (IDentification), and broadcast data. The tuner unit 254 extracts the name of the broadcast channel included in the broadcast signal, the service ID, and the broadcast data and transmits the extracted broadcast channel signal to the broadcast signal processor 256 and the controller 212.

The broadcast signal processing unit 256 divides the broadcast signal passed through the tuner unit 254 into a video broadcast signal and an audio broadcast signal, and performs a series of signal processing. A series of signal processing performed by the broadcast signal processing unit 256 may include analog-to-digital conversion or digital-to-analog conversion, converting the video data into a signal capable of driving the display 214, and the like.

The communication signal processing unit 258 processes the communication signal with the GPS satellite and the telematics communication signal. That is, the communication signal processing unit 258 converts the received communication signal into data for transmission to the control unit 212, or transmits data to be transmitted through the tuner unit 254 and the antenna 252 to the control unit 212, And converts the received signal into a signal of a communicable type.

The navigation database 262 includes data for implementing navigation. The navigation database 262 may be in the form of a memory card or a DVD (Digital Versatile Disc). Navigation data provided from a mobile terminal connected through a wired / wireless link (for example, CarPlay or Android Auto) may be utilized as a navigation database.

The navigation driving unit 264 forms a navigation screen on the display 214 using data provided from the navigation database 262. [ For this, the controller 212 receives navigation setting information such as a destination, a transit route, and a route form set by the driver. In addition, the control unit 212 receives the current position information of the vehicle secured through communication with the GPS satellite to implement the navigation.

The audio / video input unit 272 may be an optical disc drive. Or the audio / video input 272 may be a universal serial bus (USB) input / output device or a spare input / output terminal (aka AUX). Or audio / video input 272 may be a Bluetooth device for wireless connection with the mobile terminal. The mobile terminal connected to the audio / video input unit 272 via Bluetooth may be a mobile phone or a portable digital sound source reproducing apparatus.

The audio / video playback unit 274 allows the audio / video data input through the audio / video input unit 272 to be output to the speaker 116 or the display 214. [ For example, when the audio / video input unit 272 is an optical disc drive, the optical disc drive reads out the audio / video data recorded on the optical disc (CD / DVD / BD or the like) The audio / video reproducing unit 274 converts the audio / video data fetched by the audio / video input unit 272 into a signal capable of driving the speaker 116 or the display 214, Display 214 to allow the audio / video to be played back. The audio / video data provided from a medium other than the optical disc can be converted into a signal capable of driving the speaker 116 or the display 214 while passing through the audio / video reproducing unit 274.

The input unit 282 may be a touch screen implemented on the at least one button or display 214 provided in the AVN 100. [ The driver can select one of the composite functions of the AVN 100 through the operation of the input unit 282 and apply various settings so that an expected operation can be performed from the selected function. The voice recognition button 104 of the steering wheel 102 described above may be included in at least one button constituting the input unit 282. [

The control unit 212 is involved in the overall operation of the AVN 100 and performs necessary control. For example, in response to the operation of the voice recognition button 104, an application related to the voice recognition function of the memory 210 is driven to display an initial entry screen and output an associated voice guidance message. The control unit 212 also receives voice command information from the voice recognition processing unit 208 and generates a control command corresponding to the voice command information to control the voice command information. In addition, the control unit 212 may process the broadcast / communication signal. Video data to be transmitted to the speaker 116 or the display 214 when the audio / video data generated after the processing of the broadcast / communication signal is to be output to the speaker 116 or the display 214 So that output of audio / video data can be performed. When the driver selects the navigation function, the controller 212 controls the navigation database 262, the navigation driver 264, the display 214, and the speaker 116 so that navigation can be implemented. The control unit 212 controls the audio / video data input through the audio / video input unit 272 to be reproduced by the audio / video reproducing unit 274 and transmitted to the speaker 116 or the display 214, / Video data can be output. The control unit 212 converts the name of the broadcast channel extracted from the broadcast signal by the tuner unit 254 into text and transmits the converted text to the speech recognition processing unit 208. [

The memory 210 stores various applications executed to perform the voice recognition function, the broadcasting / communication function, the navigation function, and the audio / video function of the AVN 100, and the display data and voice data, Effect sound data and the like are stored.

The display 214 outputs a video accompanied by a voice recognition function of the AVN 100, a broadcast / communication function, a navigation function, and an audio / video function. For example, a guide screen for each function, a message, video data, and the like are output through the display 214.

The speaker 116 outputs audio accompanied by a voice recognition function of the AVN 100, a broadcast / communication function, a navigation function, and an audio / video function. For example, announcements, effect sounds, audio data, etc. for each function are outputted through the speaker 116.

Fig. 3 shows a configuration of the speech recognition processing unit shown in Fig. The speech recognition processing unit 208 shown in FIG. 3 is based on word recognition and non-specific speaker recognition. 3, the speech recognition processing unit 208 includes an analog-to-digital conversion unit 302, a speech analysis unit 304, a speech detection unit 306, a word recognition unit 308, a word standard pattern database 310, And a character-to-speech converting means 352.

The voice signal converted into the electric signal output from the microphone 106 is in analog form. For speech recognition, this analog type electrical signal must be converted into a digital type electrical signal. The analog-to-digital conversion unit 302 converts an analog type electrical signal input from the microphone 106 into a digital type electrical signal.

The voice analysis unit 304 analyzes the voice signal converted into the digital type electric signal and extracts the feature pattern. The voice analysis unit 304 divides the voice signal converted into the electric signal into frames of a predetermined size (for example, 10 ms or 30 ms) and extracts the feature patterns for each frame. The extraction of the feature pattern in the voice analysis unit 304 is performed through analysis of information such as frequency or amplitude of the voice signal.

The voice detection unit 306 detects a section in which the actual voice exists in the voice signal converted into the digital type electrical signal. When the vehicle is traveling, various types of noise are generated in the interior of the vehicle. For example, an engine sound, an exhaust sound, a wind noise, etc. of a car may enter the room and act as a noise. Also, while listening to music using the audio function of the AVN 100, music sound may act as noise in terms of speech recognition. In order to provide accurate and high voice recognition rate in such a noisy environment, it is very important for the AVN 100 used in an automobile interior to detect the beginning and end of a section in which a voice signal exists.

The word recognition unit 308 selects a word most similar to the voice signal input through the microphone 106. [ In the word standard pattern database 310, a plurality of reserved words (words with predefined meaning and usage) and characteristic patterns of respective reserved words are stored as standard patterns. The word recognition unit 308 compares the feature pattern acquired through the analysis of the speech signal with the standard pattern previously prepared, and selects a standard pattern most similar to the feature pattern. The word recognizing unit 308 provides the control unit 212 with the information of the reserved words corresponding to the selected standard pattern as a result of recognition.

The text-to-speech converting means 352 receives the text data of the name of the new broadcasting channel secured by the tuner unit 254 through the control unit 212, and converts the voice signal of the name of the new broadcasting channel through the text- . The voice signal of the name of the new broadcasting channel generated by the character-voice converting means 352 is transmitted to the voice analyzing unit 304 to analyze the characteristic pattern. In order to register the new broadcast channel in the word standard pattern database 310 of the speech recognition processing unit 208, a characteristic pattern is extracted by analyzing the voice signal of the name of the new broadcasting channel, and the characteristic pattern is stored in the word standard pattern database 310 You must register. This process is performed by the control unit 212 and the voice recognition processing unit 208. The first method is to obtain the voice signal of the name of the broadcasting channel by inducing the driver's utterance. In this manner, the driver's utterance is induced to secure the voice signal of the name of the broadcasting channel, so that the voice signal can be secured through the direct speech of the driver, so that the characteristic pattern of the voice signal can be extracted more accurately. In the case of securing the voice signal through the direct speech of the driver, since the driver's participation is absolutely necessary, the letter-to-speech conversion means 352 can be used as another method excluding the participation of the driver (i.e., There is a method of securing a voice signal of the name of the broadcast channel and extracting the feature pattern. In the case of the method using the text-to-speech conversion, the driver's convenience can be achieved because the driver does not need to participate.

In speech recognition, the accuracy of comparison between a feature pattern of a speech signal and a standard pattern prepared in advance greatly affects the speech recognition rate. There are many ways to compare feature patterns and comparison patterns. For example, the following two methods can be used.

Dynamic Programming Matching (DPM), a mainstream method in the 1980s, is based on the fact that the length of a word is partially changed by a vocal situation or a speed of speech even with the same word, .

Recently, the mainstream 'Hidden Markov Model (HMM)' method is a method that can cope with not only time axis change but also spectrum change by individual difference through statistical technique. The HMM prepares a pattern of each basic unit of speech in advance and statistically processes the characteristic pattern of the speech signal to determine which pattern is closest to the prepared pattern. With the development of the HMM method, the speech recognition rate of the speaker independent recognition is greatly improved. Speaker independent recognition is a technology for recognizing the speaker of an unspecified speaker, that is, an arbitrary speaker, in which pattern information of voice signals of a plurality of unspecified speakers is extracted in advance and constructed and operated as a database so that speech recognition targeting unspecified speakers is possible will be.

4 is a diagram illustrating a method of controlling speech recognition of a vehicle according to an embodiment of the present invention. A method of controlling voice recognition of an automobile according to an embodiment of the present invention includes acquiring information of a new broadcast channel for registration of a new broadcast channel and acquiring a feature pattern of a voice signal generated by speech of a name of a new broadcast channel And registering the information of the new broadcast channel and the feature pattern of the voice signal by voice recognition control. 4, the control unit 212 determines whether it is necessary to register the new broadcast channel received by the tuner unit 254 of the AVN 100 for speech recognition control (402). For example, the control unit 212 may determine that registration of a new broadcast channel for voice recognition control is necessary in the following cases.

In the first example, when the frequency of the same broadcast channel is changed by inter-area movement, the controller 212 may determine that registration of a new broadcast channel for voice recognition control is necessary. In case of SBS Power FM, the frequency of Seoul / Gyeonggi area is 107.7MHz and the frequency of Busan area is 99.9MHz. Therefore, it is impossible to receive normal broadcasting signals when the car enters the Busan area from the Seoul / Gyeonggi area to the Busan area with the frequency set to Seoul / Gyeonggi area. At this time, if the feature pattern of the voice signal of the name of the SBS power FM in the Pusan area, the service ID, and the name of the broadcast channel is not registered in the voice recognition processor 208, the driver can use the voice recognition control to control the SBS power FM I can not tune in. In this case, the control unit 212 may determine that registration of the SBS power FM in the Busan area is necessary for voice recognition control. The controller 212 receives a signal of 99.9 MHz which is the frequency of the SBS power FM of the Busan area through the antenna 252 but does not have information on the corresponding broadcast channel, It can be understood that it is not registered in the database.

As another example, when a new broadcasting station is opened and a broadcasting channel is added, the controller 212 may determine that registration of a new broadcasting channel for voice recognition control is necessary. For example, when the SBS power FM is opened at the time of production of the automobile and the SBS love FM is further opened after leaving the automobile, the voice recognition processing unit 208 of the AVN 100 of the automobile receives the SBS love FM Since the broadcast channel is not registered, the driver can not tune into the SBS Love FM using the speech recognition control. In this case, the control unit 212 may determine that registration of the SBS power FM in the Busan area is necessary for voice recognition control. Since the signal of 103.5 MHz, which is the frequency of the new broadcast channel SBS Love FM, is received through the antenna 252 but there is no information on the corresponding broadcast channel, the control unit 212 determines that the corresponding broadcast channel is the speech recognition processing unit of the AVN 100 208 that the user is not registered.

As another example, when the name of an existing broadcast channel that has already been registered is changed, the controller 212 may determine that registration of a new broadcast channel for voice recognition control is necessary. For example, if any one of the channels of the DMB TV broadcasting ends the broadcasting service and the broadcasting channel of another name is started on the same channel, the existing name and the new name of the already registered broadcasting channel do not coincide with each other, Can not tune to the broadcast channel of the new name by using the voice recognition control. In this case, the controller 212 may determine that the name of a new broadcast channel is required for voice recognition control.

For this reason, if it is determined that registration of a new broadcast channel is required (Yes in step 402), the control unit 212 acquires information on a new broadcast channel for voice recognition control and registers it in the voice recognition processing unit 208 ). It has been described that the characteristic pattern of the voice signal of the name of the new broadcasting channel can be acquired and registered through the tuner unit 254 and the voice recognition processing unit 208 in the description of FIG. 2 and FIG. According to this method, the controller 212 acquires the information of the new broadcast channel from the received broadcast signal, and registers and stores the new broadcast channel information so that the driver can later select the new broadcast channel through voice recognition control.

When the registration of the new broadcast channel is completed, the driver can select a new broadcast channel of the AVN 100 using voice recognition control (406). That is, similar to the case of the general voice recognition control, the control unit 212 of the AVN 100 receives the voice signal of the name of the desired broadcast channel from the driver, extracts the characteristic pattern of the received voice signal, It is possible to control the tuner unit 254 for tuning the corresponding broadcast channel according to whether or not the standard pattern is matched.

5 is a diagram illustrating a method of acquiring and registering information on a new broadcast channel for voice recognition control of an automobile according to an embodiment of the present invention. The method of acquiring and registering the new broadcast channel shown in Fig. 5 is the same as the method of acquiring and registering the new broadcast channel for voice recognition control in the process of determining whether registration of a new broadcast channel is necessary in the voice recognition control method of the car shown in Fig. And the registration process.

5, the controller 212 extracts the name of the broadcast channel and the service ID from the broadcast signal received by the tuner unit 254 (502). For example, when the name of the new broadcast channel is " myHYUNDAI ", and the service ID is " MH0980 ", the name and service ID of the new broadcast channel can be extracted from the broadcast signal.

The control unit 212 compares the extracted name of the new broadcast channel and the service ID with the name and service ID of the existing broadcast channel already registered in the speech recognition processing unit 208 (504). This comparison is for identifying whether the new broadcast channel is actually a new broadcast channel or a broadcast channel already registered in the speech recognition processing unit 208. [

If the name and service ID of the new broadcast channel do not match the name and service ID of the existing broadcast channel already registered in the speech recognition processing unit 208 (Yes in 506), the control unit 212 determines that the new broadcast channel It is determined that it is an actual new broadcast channel that is not registered in the speech recognition processing unit 208. [ Prior to the registration of the new broadcast channel, the control unit 212 outputs a query message for confirming the registration intention of the new broadcast channel from the driver (508). A method of outputting a query message for confirming the registration intention of a new broadcast channel from an operator is as follows.

And display the query message via the display 214. [ For example: "A new broadcast channel has been detected. Do you want to register a new broadcast channel for voice recognition control? "Is displayed on the display 214 and a selection button" Registration "and" Reject "are displayed on the display 214 as well. Forming the select button "Register" and "Deny" on the display 214 is the case when the display 214 supports the touch screen and if the display 214 does not support the touch screen, A message designating another physical button provided on the input unit 282 of the display unit 214 may be displayed on the display 214 to induce the driver to selectively operate the corresponding button. If the display 214 supports the touch screen, the control unit 212 receives a driver's acceptance / rejection request for registration of the new broadcast channel through the touch screen of the display 214 (510).

And can output a query message via the speaker 116. [ For example: "A new broadcast channel has been detected. Do you want to register a new broadcast channel for voice recognition control? If you want to register, press and hold the voice recognition button and say registration ", and if you do not want to register, press and hold the voice recognition button and say" reject ". In response to this, if the driver speaks 'registration' or 'rejection' while holding the voice recognition button 104, the driver's registration acceptance / rejection indication can be received in the form of a voice signal through the microphone 106. The received voice signal is recognized by the voice recognition processing unit 208 and the recognition result is transmitted to the control unit 212. [

When the driver touches the 'register' button on the display 214 or speaks 'registration' when the voice recognition button 104 is pressed, the controller 212 recognizes that the driver wants to register the new broadcast channel. If the driver wants to register a new broadcast channel (YES in 510), the controller 212 controls the voice recognition processor 208 to perform a registration process of a new broadcast channel (512). For this, the control unit 212 acquires a voice signal corresponding to the name of the new broadcast channel, extracts a feature pattern of the voice signal through analysis of the secured voice signal, And registers and stores the characteristic pattern of the signal in the AVN 100 as information of the new broadcast channel.

When the registration of the new broadcast channel is completed, the AVN 100 enters the standby state (514).

If the name of the new broadcast channel and the service ID match the name and service ID of the existing broadcast channel already registered in the speech recognition processing unit 208 ('No' in 506), the control unit 212 proceeds to step 506, It is determined that the new broadcast channel is an existing broadcast channel registered in the voice recognition processing unit 208 and the AVN 100 enters the standby state.

If the driver rejects the registration of the new broadcast channel (No in step 510), the controller 212 determines that the driver does not have a will to register the new broadcast channel and the AVN 100 is in the standby state Let it enter.

6 and 7 are diagrams illustrating a method of registering a new broadcast channel of a vehicle according to an embodiment of the present invention. That is, FIGS. 6 and 7 illustrate the specific method of the new broadcast channel registration process 512 of FIG. 5 described above.

6 is a diagram illustrating a first embodiment of a method of registering a new broadcast channel according to an embodiment of the present invention. The first embodiment shown in Fig. 6 is a case of acquiring a voice signal of a name of a new broadcasting channel through direct speech by a driver.

For this, the controller 212 outputs a message requesting the input of a voice signal having the name of the new broadcast channel (602). A method of outputting a message requesting a voice signal input of a name of a new broadcast channel will be described as an example.

The display 214 may output a text requesting a voice signal input of the name of the new broadcast channel. For example, the text " please say myHYUNDAI " while holding the speech recognition button to register the speech recognition of the new broadcast channel can be displayed on the display 214. [ The driver can generate a voice signal corresponding to the name of the new broadcast channel by saying 'myHYUNDAI' while holding the voice recognition button 104 in accordance with this instruction.

And can output a request for requesting the input of a voice signal of the name of the new broadcast channel through the speaker 116. For example, the phrase " please say myHYUNDAI " while holding the voice recognition button for voice recognition registration of a new broadcast channel can be output through the speaker 116. [ The driver can generate a voice signal corresponding to the name of the new broadcast channel by saying 'myHYUNDAI' while holding the voice recognition button 104 in accordance with this instruction.

The control unit 212 monitors whether the driver speaks the name of the new broadcast channel together with the operation of the voice recognition button 104 in response to the display of the request message via the display 214 or the output of the request via the speaker 116 (604). If the driver speaks 'myHYUNDAI', which is the name of the new broadcasting channel, and the voice signal is received through the microphone 106 ('Yes' in step 604), the controller 212 controls the voice recognition processor 208, And extracts a feature pattern through analysis of the speech signal (606). Then, the control unit 212 stores and registers the text of the name of the new broadcast channel, the service ID, and the characteristic pattern of the voice signal in the voice recognition processing unit 208 (608). When the broadcasting channel is selected through the voice expression of the driver according to the storage and registration of the new broadcasting channel information, the controller 212 generates a channel selection control command for the corresponding broadcasting channel in the tuner unit 254, (254) to receive the broadcast signal of the corresponding broadcast channel.

If the driver does not respond to the voice input request of the name of the new broadcast channel ('No' in 604), the controller 212 determines whether the driver has previously set the name of the new broadcast channel If the preset time elapses after waiting time elapses, the registration process of the new broadcast channel is forcibly terminated (610). If the driver does not respond to the voice input request of the name of the new broadcast channel, the driver is not ready to speak the name of the new broadcast channel or is not in an environment where the name of the new broadcast channel can be uttered In this case, the control of the controller 212 is not stopped in an infinite loop by forcibly terminating the registration process of the new broadcast channel after a predetermined time elapses rather than waiting indefinitely for the utterance of the driver, Thereby preventing the occurrence of errors in the registration of the new broadcast channel due to the noise that is erroneously entered into the channel.

7 is a diagram illustrating a second embodiment of a method of registering a new broadcast channel according to an embodiment of the present invention. The second embodiment shown in FIG. 7 is a case of converting the text of the name of the new broadcast channel into speech through the character-to-speech conversion means to acquire the speech signal of the name of the desired new broadcast channel.

For this, the controller 212 performs character-to-speech conversion of the name of the new broadcast channel using the text-to-speech conversion means (702). That is, the control unit 212 controls the text-to-speech conversion unit 252 of the speech recognition processing unit 208 to convert the text of the name of the new broadcast channel into a speech signal. When the voice signal of the name of the new broadcast channel is generated by the character-voice conversion means 252, the control unit 212 controls the voice recognition processing unit 208 to extract the feature pattern through analysis of the voice signal (704 ). Then, the control unit 212 stores and registers the text of the name of the new broadcast channel, the service ID, and the feature pattern of the voice signal in the voice recognition processing unit 208 (706). When the broadcasting channel is selected through the voice expression of the driver according to the storage and registration of the new broadcasting channel information, the controller 212 generates a channel selection control command for the corresponding broadcasting channel in the tuner unit 254, (254) to receive the broadcast signal of the corresponding broadcast channel.

The control unit 212 may display the registration of the new registration channel through the display 214 while the registration process of the new broadcast channel according to the second embodiment as shown in FIG. 7 is in progress. In the case of the first embodiment (Fig. 6) in which the voice signal of the name of the new broadcast channel is secured by the driver's direct utterance, since the driver is engaged, the driver can recognize that the registration process of the new broadcast channel is proceeding have. However, as in the second embodiment (FIG. 7), the process of securing the voice signal of the name of the new broadcast channel through the text-to-speech conversion is not visually evident, so that during the registration of the new broadcast channel, An unresponsiveness may cause the driver to be curious. Also, since the AVN 100 does not respond to the operation of the driver during the registration of the new broadcast channel, the driver can mistake the AVN 100 as a malfunction. Accordingly, during the registration process of the new broadcast channel, the control unit 212 can display that the registration of the new registration channel is currently proceeding through the display 214, thereby preventing the driver from being curious or mistaken as a failure .

8 is a diagram illustrating a method of selecting a broadcast channel through voice recognition after updating a new broadcast channel for voice recognition control of an automobile according to an embodiment of the present invention. A method of selecting a broadcast channel through speech recognition shown in FIG. 8 is a concrete process for selecting a new broadcast channel through speech recognition in the speech recognition control method of the car shown in FIG.

As shown in FIG. 8, in a state in which the AVN 100 is waiting (802), the control unit 212 monitors whether a voice signal for voice recognition control is inputted through the microphone 106 (804). That is, the control unit 212 monitors whether or not utterance through the operation of the driver's voice recognition button 104 occurs. If no audio signal is input ('No' in 804), the controller 212 keeps the AVN 100 in a standby state.

If an input of a voice signal occurs (Yes at 804), the control unit 212 performs voice recognition and pattern comparison (806). That is, when the driver speaks while pressing the voice recognition button 104 to use the voice recognition control function, the voice recognition processing unit 208 receives the voice signal of the driver through the microphone 106. [ The voice recognition processing unit 208 extracts a feature pattern of the voice signal through analysis of the voice signal of the received voice signal. In addition, the speech recognition processing unit 208 compares the extracted characteristic pattern of the speech signal with the reference pattern stored in the word standard pattern database 310. If a matching reference pattern exists, . For example, whether the reserved word of the reference pattern is an instruction for controlling the operation or selecting one of the complex functions such as radio / audio / video / DMB / navigation, or selecting a specific broadcast channel of the radio / DMB .

When the reserved word of the reference pattern is the name of the broadcast channel (Yes in 808), the control unit 212 transmits the corresponding channel information to the tuner unit 254 (810). The tuner unit 254 receiving the broadcast channel information selects the corresponding broadcast channel, receives the broadcast signal, and outputs audio / video data corresponding thereto. That is, the broadcast signal received by the tuner unit 254 is converted into audio / video data by the broadcast signal processing unit 256 located at the rear end of the tuner unit 254, and is output. The output audio / video data is displayed in the form of audio and video through speaker 116 and display 214, respectively (812).

If the reserved word of the reference pattern is not the name of the broadcast channel ('No' in 808), the control unit 212 controls related components to execute the corresponding command (814).

100: AVN (Audio / Video / Navigation)
102: Steering wheel
104: Voice recognition button
106: microphone
116: Speaker
208: Speech recognition processor
210: memory
212:
214: Display
218: Command Output Interface
252: antenna
256: Broadcast signal processor
258: Communication signal processor
262: Navigation database
264: Navigation drive unit
272: Audio / video input
274: audio / video reproducing unit
282:
302: Analog-to-digital conversion section
304: Voice recognition processor
306:
308: Word recognition unit
310: Word standard pattern database
352: Text-to-speech conversion means

Claims (22)

Obtaining information of the new broadcast channel for registration of a new broadcast channel;
Obtaining a feature pattern of a speech signal generated by speech of a name of the new broadcast channel;
And registering the information of the new broadcast channel and the feature pattern of the voice signal for voice recognition control. 2. The method of claim 1, wherein the obtaining of the feature pattern of the speech signal comprises:
Inducing the driver to speak the name of the new broadcast channel;
And analyzing the speech signal generated by the driver's speech to obtain the feature pattern. 3. The method according to claim 2,
And displaying a guidance message for inducing utterance of the name of the new broadcast channel of the driver. 4. The method of claim 3,
And displaying a guidance text for inducing utterance of the name of the new broadcast channel through a display. 4. The method of claim 3,
And outputting an announcement for inducing utterance of the name of the new broadcast channel through a speaker. 2. The method of claim 1, wherein the obtaining of the feature pattern of the speech signal comprises:
Converting a character string of the name of the new broadcast channel into a voice signal using a character-voice conversion means;
And analyzing the speech signal generated by the character-to-speech conversion to obtain the feature pattern. The method according to claim 1,
Wherein the information of the new broadcast channel includes a name of the new broadcast channel and a service ID. A tuner unit for acquiring information of the new broadcast channel for registration of a new broadcast channel;
A speech recognition processor for obtaining a feature pattern of the speech signal from a speech signal generated by speech of the name of the new broadcast channel;
And a controller for registering the information of the new broadcast channel and the feature pattern of the voice signal for voice recognition control. The speech recognition apparatus according to claim 8,
Guiding the driver to speak the name of the new broadcast channel and analyzing the speech signal generated by the driver's speech to obtain the feature pattern. 10. The apparatus according to claim 9,
The guidance message for inducing the driver to speak the name of the new broadcast channel is displayed. The method according to claim 10,
And displays a guidance text for inducing a speech of the name of the new broadcast channel through a display. The method according to claim 10,
And outputs the announcement for inducing utterance of the name of the new broadcast channel through the speaker. The speech recognition apparatus according to claim 8,
Converting the character string of the name of the new broadcast channel into a speech signal by using a character-to-speech conversion means, and analyzing the speech signal generated by the character-to-speech conversion to obtain the characteristic pattern. 9. The method of claim 8,
Wherein the information of the new broadcast channel includes the name of the new broadcast channel and the service ID. Obtaining information of the new broadcast channel for registration of a new broadcast channel;
Displaying a guide message to induce a name of the new broadcast channel to be ignited;
Obtaining a feature pattern of a speech signal generated by speech of a name of the new broadcast channel;
And registering the information of the new broadcast channel and the feature pattern of the voice signal for voice recognition control. 16. The method of claim 15,
And displaying a guidance text for inducing utterance of the name of the new broadcast channel through a display. 16. The method of claim 15,
And outputting an announcement for inducing utterance of the name of the new broadcast channel through a speaker. A tuner unit for acquiring information of the new broadcast channel for registration of a new broadcast channel;
A speech recognition processor for obtaining a feature pattern of the speech signal from the speech signal;
Controls the output unit to display a guidance message for inducing the name of the new broadcast channel to be displayed and controls the speech recognition processor to acquire a feature pattern of a speech signal generated by utterance of the name of the new broadcast channel,
And registering the information of the new broadcast channel and the characteristic pattern of the voice signal in the voice recognition processing unit. The method as claimed in claim 18,
And displays a guidance text for inducing a speech of the name of the new broadcast channel through a display. The method as claimed in claim 18,
And outputs the announcement for inducing utterance of the name of the new broadcast channel through the speaker. Obtaining information of the new broadcast channel for registration of a new broadcast channel;
Converting a character string of the name of the new broadcast channel into a voice signal using a character-voice conversion means;
Obtaining a characteristic pattern of the speech signal generated by the character-to-speech conversion;
And registering the information of the new broadcast channel and the feature pattern of the voice signal for voice recognition control. A tuner unit for acquiring information of the new broadcast channel for registration of a new broadcast channel;
A speech recognition processor for converting a character string of the name of the new broadcast channel into a speech signal using character-to-speech conversion means and acquiring a characteristic pattern of the speech signal from the speech signal;
And registering the information of the new broadcast channel and the characteristic pattern of the voice signal in the voice recognition processing unit.

Images